Safety concerns surrounding skin-lightening agents have intensified following chemical leukoderma linked to rhododendrol. Here, we performed an in vitro safety and hazard profiling comparison of raspberry ketone (RK) and a total anthraquinone fraction from Fallopia multiflora var. cillinerve (Polygonum cillinerve) using an immortalized keratinocyte–melanocyte co-culture model (human HaCaT keratinocytes and murine B10.BR melanocytes, 3:1). Rhododendrol and arbutin were included as contextual references. Following viability-guided range finding, cells were exposed for 48 h and evaluated for melanocyte stress and injury, including ROS generation, UPR/ER-stress activation (PERK/eIF2α–ATF4-associated readouts: ATF4, Hmox1, GADD45a; and IRE1 phosphorylation), IL-8-related chemokine output (CXCL1/KC, a murine functional homolog of IL-8), cell-cycle perturbation, and Caspase-3-associated apoptosis. In parallel, targeted LC–MS metabolomics was performed to resolve pathway-level perturbations. High-dose RK elicited a rhododendrol-like in vitro stress/toxicity signature, characterized by elevated ROS, robust UPR engagement, inflammatory chemokine induction, cell-cycle dysregulation, and pro-apoptotic responses; under viability-adjusted conditions, these effects remained more evident than with arbutin. Metabolomics revealed convergent disturbances between RK and rhododendrol, highlighting purine metabolism as a prominent perturbed pathway and suggesting purine-related metabolites as candidate indicators associated with leukoderma-relevant cellular stress in vitro. In contrast, the anthraquinone fraction did not trigger oxidative or ER stress within the tested range and exhibited a more favorable in vitro safety profile, including reduced ROS.
Hou et al. (Sat,) studied this question.